Quarterisation of national income accounts of Pakistan

Munich Personal RePEc Archive

Quarterisation of national income accounts of Pakistan

Hanif, Nadim and Iqbal, Javed and Malik, Jehanzeb

State Bank of Pakistan

21 March 2013

Online at https://mpra.ub.uni-muenchen.de/45334/

MPRA Paper No. 45334, posted 21 Mar 2013 12:35 UTC

STATE BANK OF PAKISTAN

March, 2013 No. 54

M. Nadim Hanif Javed Iqbal M. Jahanzeb Malik

Quarterisation of National Income Accounts of Pakistan

SBP Working Paper Series

SBP Working Paper Series

Editor: Dr. Hamza Ali Malik

The objective of the SBP Working Paper Series is to stimulate and generate discussions, on different aspects of macroeconomic issues, among the staff members of the State Bank of Pakistan. Papers published in this series are subject to intense internal review process. The views expressed in the paper are those of the author(s) and do not necessarily reflect those of the State Bank of Pakistan.

© State Bank of Pakistan, All rights reserved.

Price per Working Paper

Pakistan: Rs 50 (inclusive of postage) Foreign: US$ 20 (inclusive of postage)

Purchase orders, accompanied with cheques/drafts drawn in favor of State Bank of Pakistan, should be sent to:

Chief Spokesperson

External Relations Department, State Bank of Pakistan,

I.I. Chundrigar Road, P.O. Box No. 4456, Karachi 74000. Pakistan

For all other correspondence:

Editor,

SBP Working Paper Series Research Department, State Bank of Pakistan,

I.I. Chundrigar Road, P.O. Box No. 4456, Karachi 74000. Pakistan

Published by: Editor, SBP Working Paper Series, State Bank of Pakistan, I.I. Chundrigar Road, Karachi, Pakistan.

ISSN 1997-3802 (Print) ISSN 1997-3810 (Online)

http://www.sbp.org.pk

Printed at the SBPBSC (Bank) – Printing Press, Karachi, Pakistan

Quarterisation of National Income Accounts of Pakistan

M. Nadim Hanif, Additional Director, State Bank of Pakistan, Nadeem.Hanif@sbp.org.pk Javed Iqbal, Joint Director, State Bank of Pakistan, Javed.Iqbal6@sbp.org.pk

M. Jahanzeb Malik, Joint Director, State Bank of Pakistan, Jahanzeb.Malik@sbp.org.pk

Abstract

Arby (2008) quarterised the production side of annual GDP, and its subsectors, for 1972 to 2005 based on constant prices of 1999-2000 as well as on current prices. This study provides quarterly estimates of (sectoral and overall) gross domestic production in Pakistan during 1999-2000 to 2009- 2010 based on constant prices of 1999-2000 as well as on current prices. Seasonality in quarterly gross domestic product in Pakistan mainly arises from agriculture. Furthermore, the study also provides estimates of various components of expenditures side of the GDP for 1972-1973 to 2009- 2010 for the first time as far as we know.

Key words: Quarterisation, National Income Accounts JEL Classification: E01, Y10

Acknowledgment

Authors would like to thank Ali Choudhary, Behzad Ali Ahmed, Farooq Arby, Farooq Pasha and Umar Siddique for their helpful comments on the earlier draft of the paper. Special thanks to Imran Naveed Khan for helping estimation of expenditure side of this study. Any errors or omissions are the responsibility of the authors. Finally, views expressed here are those of the authors and not necessarily of the State Bank of Pakistan.

Contact of author for correspondence Name: Dr. M. Nadim Hanif

Designation: Additional Director Department: Research

State Bank of Pakistan

I.I. Chundrigar Road

Karachi - 7400

Pakistan

Non-technical Summary

National Income Accounts (NIA) in Pakistan are regularly compiled by Pakistan Bureau of Statistics (PBS) on annual basis exclusively. PBS does not disseminate the NIA on quarterly basis which could provide a timelier picture than their annual counterpart.

Efforts have been made in the past by the researchers on the quarterisation of annual NIA of Pakistan (like Bengaliwala (1995), Kemal & Arby (2004) etc.). The most recent of such efforts is (Arby, 2008). Though, Arby (2008) provides the quarterly estimates of GDP and its production side sectors/subsectors (for 1972 to 2005) based on current prices and on constant prices of 1999-2000, it lacks production side quarterly estimates of GDP for the second half of the last decade and is silent over the quarterly estimates of expenditure side components of GDP.

We provide estimates of quarterly GDP from production side (both on current prices and on constant prices of 1999-2000) from 2000 to 2010 (the year up to which we have all desired data, necessary for quarterly NIA estimation) following Arby (2008). Combining quarterly GDP production side estimates for 1973-1999 based upon Arby (2008) and those for 2000 to 2010 from this study, we get quarterly GDP from production side for 1973 to 2010. Once we obtain quarterly GDP from production side for 1973 to 2010, we quarterise the expenditures side of the NIA. Quarterly estimates of overall gross fixed capital formation are worked out following Arby and Batool (2007). Government fixed capital formation is quarterised on the basis of quarterly proportion of government total development expenditures. Exports and imports are quarterised using monthly trade data of PBS. After quarterisation of government consumption expenditure on the basis of gross public services expenditure, private consumption expenditures are estimated as a residual.

These data will be useful for business cycle analysis and economic modeling. Of

course, for understanding the economy’s structure and long term trends, annual NIA are more

useful.

1 1. Introduction

National Income Accounts (NIAs) present macro picture of income cycle of a country. It helps us know the sources of supply of goods and services (agriculture, industry and services) and the uses of the same (final consumption expenditures, gross capital formation, and exports).

In Pakistan, Pakistan Bureau of Statistics (PBS) compiles Annual National Income Accounts (ANIA) based on UN System of National Accounts 1993

. Some information is also available on monthly basis like large scale manufacturing (LSM) index

. PBS does not disseminate Quarterly National Income Accounts (QNIA) which can provide timelier picture than ANIAs and more comprehensive than provided by the indicators like LSM index and merchandise imports/exports.

This study provides the estimates of the quarterly GDP from production as well as expenditures side (both on current prices and on prices of 1999-2000) and its various subsectors/components since 1972-73 following Arby (2008), and Arby and Batool (2007) by using available data.

QGDP production side estimates for 1973-1999 are based upon Arby (2008) while those for 2000-2010 are going to be worked out in this study following Arby (2008). Once we obtain QGDP from production side for 1973 to 2010 we will quarterise the expenditures side of the NIA. Quarterly estimates of overall gross fixed capital formation will be worked out following Arby and Batool (2007). Government fixed capital formation will be quarterised on the basis of quarterly proportion of government total development expenditures. Net exports will be quarterised using monthly trade data of PBS. After quarterisation of government consumption expenditures on the basis of gross public services expenditures, private consumption expenditures will be estimated as a residual.

With these datasets, researchers will be able to do better analysis of dynamic relationships between these aggregates and between these and other important aggregates (like broad money, consumer prices, etc.). Thus, these data will be useful for business cycle analysis and economic modeling. Of course, for understanding the economy‟s structure and long term trends ANIA are more useful

. Though, Arby (2008) provides the quarterly estimates of GDP and its production side sectors/subsectors (for 1972 to 2005) based on current prices and on constant prices of 1999-2000, it lacks production side quarterly estimates of GDP for the second half of the last decade and is silent over the quarterly estimates of expenditure side components of GDP.

There is need to provide the production estimates at least for the second half of 2010s.

Since there are usually revisions in the NIA upto 3 years from the first release of the data, there

1 The 1993 System of National Accounts (SNA 1993) is a comprehensive, consistent and flexible set of macroeconomic accounts to meet the needs of government and private sector analysts, policy makers and decision takers. It is a conceptual framework, not manual for compilations of NIAs. For that purpose a separate set of handbooks like „Handbook of National Accounting,‟ is published by United Nations. SNA 1993 was prepared by Inter-Secretariat Working Group of National Accounts (ISWGNA) - which consists of IMF, EU, OECD, UN and WB – mandated by the Statistical Commission of United Nations to oversee the international coordination on the development of national accounts. Based on evolving needs of users, new developments in economic environment, and advances in methodological research ISWGNA has produced and released updated version SNA 2008. We would like PBS to compile and disseminate national income accounts on the basis of SNA 2008 as soon as possible.

SNA 2008 is designed to accommodate the needs of countries at different stages of economic development.

2 It is an index of the production of large scale manufacturing establishments in Pakistan registered under Factories Act, 1934 (for details see PBS, 2011). It is based upon production of 112 commodities as reported regularly in Monthly Statistical Bulletin of PBS. The weights in the construction of the index are based upon census of manufacturing industries (CMI) conducted in the country from time to time. Recent most weights are from PBS (2009).

3 For example, changes in the share of agriculture in overall income changes in decades and not in quarters.

2

is also a need to reassess the quarterly GDP even from 1999-2000. We have estimated quarterly GDP from production side from 1999-2000 to 2009-2010 (the year up to which we have all desired data, necessary for quarterly NIA estimation). We also provide the quarterly estimates of various components of expenditures side of the NIAs for FY1973-FY2010.

The remaining of this paper is organised as follows. Section 2 presents some review of literature. Section 3 describes the methodology to estimate the quarterly NIA (sector- wise as well as sub-sector wise) from production side. In section 4 we discuss (component wise) the approach adopted to estimate the QNIA from expenditures side. The estimates of quarterly NIA (overall as well as component-wise) are presented in appendices: Appendix A presents time series of quarterly GDP (Real) from production side for 1999-00 to 2009-10. Appendix B presents time series of quarterly GDP (Nominal) from production side for 1999-00 to 2009-10.

In these appendices (A and B) sector and subsector wise quarterly estimates are provided from production side worked in this study. Appendix C combines the quarterly proportion for GDP(FC) obtained in Arby (2008) for 1972-73 to 1998-99 and those estimated in this study for 1999-00 to 2009-10. For combined period of 1972-73 to 2009-10 we have provided current as well real quarterly GDP(FC) in Appendix D. Appendices E1 to E11 present quarterly estimates of various components of expenditures side of GDP. Appendix E12 contains quarterly estimates of net factor income from abroad. Appendices F, G and H present QGD(MP), QGNP(FC) and QGNP(MP) respectively. In section 5 some observations presented based on the quarterly estimates of GDP are analyzed. The last section presents concluding remarks.

2. Literature Review

The past efforts on the quarterisation of ANIA of Pakistan include a recent study (Arby, 2008) on the issues of NIAs of Pakistan. It provides a very good account of the relevant literature. Earlier studies on the estimates of Pakistan‟s QNIAs include Bengaliwala (1995), Kemal and Arby (2004), and Arby (2008). Bengaliwala (1995) and Kemal and Arby (2004) quarterise old base (1980-81) GDP data and these studies considered only the contact prices and production side of NIAs. None of these studies provide the latest estimate for quarterly production in Pakistan, which we have provided in this study (up to Q4-FY2010)

Arby and Batool (2007) quarterised the overall gross fixed capital formation for 1972 to 2006. However, our study provides the quarterly private and public sector gross fixed capital formation separately in addition to overall gross fixed capital formation. This study attempts to fill this gap of previous studies by providing quarterly estimates of various components of expenditures side of the NIAs for 1972-73 to 2009-10.

3. Methodology-Production Side

For quarterisation of production side we have followed the methodology used in Arby (2008) and re-estimated the quarterly production for the period of 1999-00 to 2009-10. In the following subsections we discuss the quarterisation process for different sectors and subsectors of the economy.

3.1 Agriculture Sector

This sector consists of five subsectors, major crops, minor crops, live stocks, fishing and

forestry. How quarterly estimates of gross value added of these sub-sectors (for FY00 to FY10)

are obtained is detailed below.

3 3.1.1 Major Crops

The value added of 12 major crops (VAMC)

has been quarterised ( 𝑉𝐴𝑀𝐶

𝑐 ; 𝑐 = 1,2, … ,12, ; 𝑡 𝑖𝑠 𝑡𝑖𝑚𝑒 𝑦𝑒𝑎𝑟 𝑠𝑐𝑟𝑖𝑝𝑡 𝑎𝑛𝑑 𝑞 = 1,2,3,4) using

a) Province-wise each major crop production ( 𝑃𝑅𝑀𝐶

𝑝 , 𝑐 ; 𝑝 = 1,2,3,4 𝑎𝑛𝑑 𝐶 = 1,2, … ,12) taken from Agriculture Statistics of Pakistan (various isssues)

b) Share of each major crop in annual value added (calculated from information (a) above) of major crops in Pakistan ( 𝑆

; 𝑐 = 1,2, … ,12;

𝑆

= 1)

c) Annual value added of major crops in Pakistan ( 𝑉𝐴𝑀𝐶

𝑐 ; 𝑐 = 1,2, … ,12) on constant prices of 1999-2000

d) Province-wise quarterly harvest calendar

of each of the 12 major crops prepared by PBS.

The quarterly value added at 1999-00 prices for quarter 𝑞 year 𝑡 is then obtained as follows. Province-wise each major crop production is converted into quarterly province-wise each major crop production ( 𝑃𝑅𝑀𝐶

𝑝 , 𝑐 ; 𝑝 = 1,2,3,4 𝑎𝑛𝑑 𝑐 = 1,2, … ,12) based upon province-wise quarterly harvest calendar. This is then summed up over provinces to get quarter- wise each major crop production, i.e.

𝑃𝑅𝑀𝐶

𝑐 =

𝑃𝑅𝑀𝐶

𝑝 , 𝑐 for 𝑐 = 1,2, … ,12

Based on the share of each major crop in annual valued added of major crops we obtain the value of each crop, i.e.

𝑉𝐴𝑀𝐶

( 𝑐 ) = 𝑆

∗ 𝑉𝐴𝑀𝐶

; 𝑐 = 1,2, … ,12;

𝑆

= 1).

This annual valued added of each major crop is then divided into four quarters based upon quarter-wise proportion of each major crop production, i.e.

𝑉𝐴𝑀𝐶

𝑐 = 𝑉𝐴𝑀𝐶

𝑐 ∗ [ 𝑃𝑅𝑀𝐶

𝑐 𝑃𝑅𝑀𝐶

𝑐

𝑞=1

]

Summing this over 12 major crops we get quarterly value added of major crops 𝑉𝐴𝑀𝐶

on constant prices of 1999-2000. Then we convert the constant prices quarterly value addition to current prices using quarterly a price index of major crops (constructed using WPI prices of relevant major crops) as data on Producer Price Index (which is used by PBS) is not (publically) available.

3.1.2 Minor Crops

The value added by 37 minor

crops has been quarterised by the same process as major crops except that harvest calendar used here is related to minor crops. We convert the constant

4 These include wheat, rice (basmati, iri and others), bajra, jowar, maize, barley, gram, sugarcane, rapeseed & mustard, and sesamum, cotton, and tobacco.

5 Pakistan Bureau of Statistics has prepared harvest calendar for each of the major (as well as minor) crop and for each of the province, on the basis of output harvested in every quarters. For example in the case of sugarcane production in Punjab, 0 percent of sugarcane of the crop is harvested in Jul-Sep, 15 percent in Oct-Dec, 80 percent in Jan-Mar and 5 percent in Apr-Jun. These proportions are 0, 38, 61 and 1 in Sindh; 0, 50, 50, and 0 in case of Khyber Pakhtonkhawah; and 0, 0, 100 and 0 in case of Baluchistan for Jul-Sep, Oct-Dec, Jan-Mar, and Apr-Jun respectively. Though such calendar has practical merit in using it for quarterisation of output, it does not consider economic activities pertaining to sowing of the crops which may have happened in some earlier quarters. Furthermore, one might not expect shocks to the harvest calendar over the short run however, over the long run there can be changes in the crops pattern across the quarters and/or provinces.

4

prices quarterly value addition to current prices using quarterly a price index of minor crops (constructed using WPI prices of relevant minor crops).

3.1.3 Live Stocks

There are two main contributors on live stocks; Milk and other. Other includes natural growth of animals, draught power, dung and urine, wool, poultry etc. Share of milk in total livestock value addition is found to be around 60 percent (or, 𝑚

= 0.60) on the basis of authors estimates based upon annual relevant statistics on milk and others. Following Kemal and Arby (2004), we can say that peak production of milk is obtained in winter. Kemal and Arby (2004) stated that 35 percent of milk production was received in summer (which they divided equally for first and fourth quarter) and 65 percent was obtained in winter season (which they divided in second and third quarter in proportion of 35:30 since third quarter is relatively warmer than the second one). The rest of the value added in the live stock is distributed equally in the four quarters. Thus,

𝑉𝐴𝐿𝑆

= 𝑉𝐴𝐿𝑆

= 𝑉𝐴𝐿𝑆

∗ (0.175 𝑚

+ 0.25 1 − 𝑚

)

𝑉𝐴𝐿𝑆

= 𝑉𝐴𝐿𝑆

∗ (0.35 𝑚

+ 0.25 1 − 𝑚

); 𝑉𝐴𝐿𝑆

= 𝑉𝐴𝐿𝑆

∗ (0.30 𝑚

+ 0.25 1 − 𝑚

).

We convert the constant prices quarterly value addition to current prices using quarterly wholesale price index of fresh milk, eggs, chicken and meat.

3.1.4 Fishing

Fishing consists of marine fishing and land (farm) fishing. Following Kemal and Arby (2004) it is assumed that marine fish makes about 30 percent of the total value added of fishing and the rest 70 percent is from land fish. Following seasonal pattern for marine and land fish are reported by PBS (2002).

Table 1: Seasonal factors of Fishing

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Marine Fish 0.2012 0.3069 0.2365 0.2554 Land Fish 0.1121 0.2735 0.3498 0.2646 Source: PBS (2002)

Thus quarterly value added of fishing is calculated as 𝑉𝐴𝐹

= 0.3 ∗ 𝑚𝑓

+ 0.7 ∗ 𝑛𝑓

∗ 𝑉𝐴𝐹

. Where 𝑉𝐴𝐹

=Value added of fishing in quarter q of year t, 𝑉𝐴𝐹

= Value added of fishing in year t, 𝑚𝑓

= Share of Marine Fish in quarter q, 𝑛𝑓

= Share of Land Fish in quarter q.

We convert the constant prices quarterly value addition to current prices using quarterly wholesale price index of fish.

6 These include mash, masoor, mung, mattar, other pulses, tomato, potato, other vegetables, groundnuts, soybean, sunflower, safflower, canola, linseed, castro seed, mango, banana, apple, citrus fruits, dates, guava, apricot, peach, pears, plums, grapes, pomegranate, almonds, chillies, onion, garlic, turmeric, ginger, Other condiments, gouar seeds, fodder crops, and sugar beet.

5 3.1.5 Forestry

Annual value added by forestry sub sector is divided into quarterly value additions using quarterly proportion 0.2009, 0.2561, 0.2882, and 0.2548 estimated by Quaidian Economic Consultants (2001). We convert the constant prices quarterly value addition to current prices using quarterly wholesale price index of timber and firewood.

The overall estimated quarterly gross value added by agriculture sector (at 1999-2000 prices) is shown below:

3.2 Industrial Sector

Industrial sector consists of six subsectors: mining and quarrying; small size manufacturing; large scale manufacturing; slaughtering; construction; and electricity, water &

gas supply. The detailed process of quarterisation of gross value added by these sub-sectors (for FY2000 to FY2010) is given below.

3.2.1 Mining and Quarrying

This subgroup consists of aragonite marble, agri-clay, silica sand, gypsum, lime stone, ordinary stone, rock salt, coal, copper, copper blister, Iron ore, bauxite, barites, bentonite, celestite, china clay, chalk, chromites, dolomite, laterite, feldspar, fire clay, fluorite, fuller‟s earth, magnesite, manganese, ochres, ordinary sand, soap stone, sulphur, erby stone, phosphate, allied services, minerals exploration establishments, natural gas, and crude oil. Monthly production of these items (except phosphate, allied services and minerals exploration establishments) is available in Monthly Statistical Bulletin (PBS). We multiply monthly production with base year prices to make quarterly gross value. Summing over the monthly gross values we calculate the quarterly gross values. Annual gross value of phosphate, allied services and minerals exploration establishments has been distributed into quarterly according to quarterly ratios of addition of items for which monthly production is available. Quarterly value

160000 210000 260000 310000 360000 410000

Q1-FY00 Q3-FY00 Q1-FY01 Q3-FY01 Q1-FY02 Q3-FY02 Q1-FY03 Q3-FY03 Q1-FY04 Q3-FY04 Q1-FY05 Q3-FY05 Q1-FY06 Q3-FY06 Q1-FY07 Q3-FY07 Q1-FY08 Q3-FY08 Q1-FY09 Q3-FY09 Q1-FY10 Q3-FY10

M i l l i o n R u p e e s

Figure 1: GVA of Agriculture - Quarterly (1999-00 Prices)

6

addition is calculated by subtracting input costs from the gross value

. We convert the constant prices quarterly value addition to current prices using composite quarterly wholesale price index of coal and natural gas (with their respective share in total mining output as weights).

3.2.2 Manufacturing

a) Large-scale Manufacturing

Annual value added by large scale manufacturing ( 𝐿𝑆𝑀

) is quarterised on the basis of quarterly shares of large scale manufacturing calculated using monthly large scale manufacturing index

(from PBS Monthly Statistical Bulletin), i.e.;

𝐿𝑆𝑀

=

3𝑘𝑚=3𝑘−2 𝐿𝑆𝑀𝑚𝑦

12𝑚=1

. 𝐿𝑆𝑀

Where 𝐿𝑆𝑀

is value added by large scale manufacturing sector in quarter q and fiscal year y and 𝐿𝑆𝑀

is monthly index of large scale manufacturing of month m and year y. We convert the constant prices quarterly value addition to current prices using wholesale price index of manufacturing.

b) Small-scale Manufacturing

Annual value added by small scale manufacturing ( 𝑆𝑆𝑀

) is also quarterised on the basis of monthly large scale manufacturing index (on the basis of backward and forward linkages between these two sectors) using

𝑆𝑆𝑀

=

3𝑘 𝑚=3𝑘−2

𝐿𝑆𝑀_𝑚𝑦

12𝑚=1

. 𝑆𝑆𝑀

Where 𝑆𝑆𝑀

is value added by small scale manufacturing sector in quarter q and fiscal year y and 𝐿𝑆𝑀

is monthly index of large scale manufacturing of month m and year y. Again, we convert the constant prices quarterly value addition to current prices using wholesale price index of manufacturing.

c) Slaughtering

Annual value added by slaughtering (both at constant and current prices) has been quarterised using shares estimated in PBS study on slaughtering industry (PBS, 2002a). These are, 18, 25, 35 and 22 percent for quarter 1, 2, 3 and 4 respectively.

3.2.3 Construction

Annual value added by construction is quarterised on the basis of (lagged) quarterly seasonal factors in cement production (taken from PBS Monthly Statistical Bulletin). That is

7 Input costs proportions (of output) of coal, crude oil & natural gas, surface minerals, allied services/mineral exploration, and other minerals are 0.24, 0.23, 0.21, 0.47 and 0.21 respectively (Pakistan Bureau of Statistics, 2004).

8 Since monthly large scale manufacturing index is based upon the actual monthly large scale manufacturing output, we can use this index for quarterisation of value added of large scale manufacturing sector.

7

𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛

= 𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛𝑉

× 𝐶𝑒𝑚𝑒𝑛𝑡

𝐶𝑒𝑚𝑒𝑛𝑡

𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛

= 𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛𝑉

× 𝐶𝑒𝑚𝑒𝑛𝑡

𝐶𝑒𝑚𝑒𝑛𝑡

𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛

= 𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛𝑉

× 𝐶𝑒𝑚𝑒𝑛𝑡

𝐶𝑒𝑚𝑒𝑛𝑡

𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛

= 𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛𝑉

× 𝐶𝑒𝑚𝑒𝑛𝑡

𝐶𝑒𝑚𝑒𝑛𝑡

Where 𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛

is the (quarterly) value added of construction in quarter k and year t and 𝑉𝐴 _ 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛

is the (annual) value added of construction in year t. Similarly, 𝐶𝑒𝑚𝑒𝑛𝑡

is the (quarterly) value added of cement in quarter k and year t and 𝐶𝑒𝑚𝑒𝑛𝑡

is the (annual) value added of cement in year t.

We convert the constant prices quarterly value addition to current prices using quarterly wage index, which is obtained by applying seasonal factors of wholesale price index of building material to annual wage index.

3.2.4 Electricity, water & gas supply

Annual value added by this sub-sector is quarterised on the basis of quarterly factors according to the study on electricity, gas & water supply (PBS, 2002b). The quarterly pattern of consumption of this group is given as;

Consumption pattern of electricity, gas & water supply sector

Group\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun

Electricity, gas & water supply 0.245 0.246 0.233 0.275 Source: PBS (2002b)

Annual value added by this sub-sector is given in different issues of Monthly Statistical Bulletin of PBS. The above factors are applied on annual value added of this sub-sector to get its quarterly estimates.

We convert the constant prices quarterly value addition to current prices using wholesale price index of „fuel and lighting‟ for electricity and gas; and general wholesale price for water.

The overall estimated quarterly gross value added of industrial sector (at 1999-00 prices) is

shown below:

8 3.3 Services

This sector consists of six subsectors: wholesale and retail trade; transport, storage, and communication; finance and insurance; ownership of dwellings; public administration, and defense and; social community, and personal services. How quarterly estimates of gross value added by these sub-sectors (for FY00 to FY10) are obtained is detailed below.

3.3.1 Wholesale and Retail Trade

PBS estimates the annual value added by this subsector by applying trade margins, which is 16 percent, on (i) imports related trade, and (ii) marketable portions of domestic production.

Annual value added by first component of this subsector is obtained as 16 percent of sum of a) consumer goods, b) 75 percent of raw material for consumer goods, and c) 55 percent of capital goods, and raw material for capital goods imports. Annual value added by second component is obtained as residual from overall value added by wholesale and retail trade. Annual value added by (i) and (ii) is quarterised on the basis of seasonal factors of imports and commodity producing sector respectively. We convert the constant prices quarterly value addition of first component to current prices using wholesale price index. However, the constant prices quarterly value addition of second component is converted to current prices using quarterly unit value index of imports (base 1999-2000).

3.3.2 Transport, Storage and Communication

This subsector covers economic activities such as transportation by railways, road, water and air; storage; pipelines for oil and gas transportation; postal services, telecom services, TV, Radio, etc. Because of the absence of quarterly (time series) information about the value addition by any of these services, we do not have any quantifiable base to quarterise the output by transport, storage and communication subs sector. However, Lisman and Sandee (1964) provide

180000 230000 280000 330000 380000 430000

Q1-FY00 Q3-FY00 Q1-FY01 Q3-FY01 Q1-FY02 Q3-FY02 Q1-FY03 Q3-FY03 Q1-FY04 Q3-FY04 Q1-FY05 Q3-FY05 Q1-FY06 Q3-FY06 Q1-FY07 Q3-FY07 Q1-FY08 Q3-FY08 Q1-FY09 Q3-FY09 Q1-FY10 Q3-FY10

M i l l i o n R u p e e s

Figure 2: GVA of Industry - Quarterly (1999-00 Prices)

9

a procedure for quarterisation of value addition of a sector/subsector where we do not have any other information set to assume the pattern of actual (quarterly fluctuations). Lisman and Sandee (1964) developed the following matrix to quarterise the annual value addition:

𝑥

𝑥

𝑥

𝑥

=

0.073 0.198 − 0.021

− 0.010 0.302 − 0.042

− 0.042 0.302 − 0.010

− 0.021 0.198 0.073

𝑋

𝑋

𝑋

Where 𝑋

is annual observation for year t, and 𝑥

is quarterly estimate for quarter j in year t

. We convert the constant prices quarterly value addition to current prices using wholesale price index of transport and communication.

3.3.3 Finance and Insurance

PBS estimates the annual value added by this subsector by summing wages, net operating surplus and depreciation pertaining to State Bank of Pakistan, scheduled and non-scheduled banks, development financial institutions (DFIs), cooperative societies, leasing, modarbas and insurance companies. Wages related part of value addition is quarterised by using the methodology given below under the “Public Administration and Defense . ” The operating surplus is quarterised by using seasonality in monetary aggregate (M2)

. Depreciation is distributed over the four quarters uniformly. We convert the constant prices quarterly value addition to current prices using consumer price index.

3.3.4 Ownership of Dwelling

Annual gross value addition of this subsector is quarterised following Lisman and Sandee (1964) method. We convert the constant prices quarterly value addition to current prices using house rent index.

3.3.5 Public Administration & Defense

PBS estimates the annual value added by this subsector based upon three components (a) wages and salaries of government employees, (b) a non-wage bill which is computed by PBS as a fixed ratio of 10% of wages, (c) and depreciation which is computed by PBS as a fixed ratio of 5% of the sum of wage and non-wage bills. While we have uniformly distributed non-wage bill and depreciation into four quarters, the wages are quarterised using the following methodology:

9 Lisman and Sandee (1964) method ensures: a) that the sum of quarterly figures equal to the yearly total; b) that if the yearly totals in three successive years are 𝑋𝑡−1, 𝑋𝑡 and 𝑋𝑡+1 , the quarterly figures during year 2 are the same but in reverse order from what they would have been had the yearly totals been 𝑋_𝑡+1 , 𝑋_𝑡 and 𝑋_𝑡−1; c) that if the yearly totals in three successive years rise by equal steps (𝑋_𝑡− 𝑋_𝑡−1=𝑋_𝑡+1− 𝑋_𝑡) the quarterly figures during year 2 also rise by equal steps; and that if 𝑋_𝑡− 𝑋𝑡−1=𝑋𝑡− 𝑋𝑡+1 the quarterly figures during year 2 should lie on sinusoid.

10 Operating surplus of financial subsector is mainly the income from assets of banks created during the period in the forms of lending which constitute part of broad money (M2).

10

In a fiscal year the wages remain constant for the first six months and then jump upward due to annual increment (usually 5 percent on average

) in January and then again remain constant for the subsequent months. The wages and salaries are quarterly distributed as under;

Let MWAS

represents the monthly wages and salaries during the first half of fiscal year t, and i represents annual increment as fraction of MWAS

. Then monthly wages and salaries for the second half will be 1 + i × MWAS

. Thus, total wages and salaries ( 𝐴𝑊𝐴𝑆

) during the year t will be:

𝐴𝑊𝐴𝑆

= 6 × ( 𝑀𝑊𝐴𝑆

+ 𝑀𝑊𝐴𝑆 × 1 + 𝑖 ) = 6 𝑀𝑊𝐴𝑆

2 + 𝑖 Fraction of wages and salaries in each quarter will be;

𝐹

= 𝐹

= 3 × 𝑀𝑊𝐴𝑆

𝐴𝑊𝐴𝑆

= 0.5/(2 + 𝑖 ) 𝐹

= 𝐹

=

𝑡

= 0.5(1 + 𝑖 )/(2 + 𝑖 ) … … … … .

The quarterly seasonal factors for Jul-Sep, Oct-Dec, Jan-Mar, and Apr-Jun are found to be 0.244, 0.244, 0.256, and 0.256 respectively.

. We convert the constant prices quarterly value addition to current prices using consumer price index.

3.3.6 Social, Community & Personal Services

Annual gross value addition of this subsector is quarterised following Lisman and Sandee (1964) method. We convert the constant prices quarterly value addition to current prices using consumer price indices for cleaning laundry & personal appearance, and recreation, entertainment & education groups combined with their respective weights in CPI basket.

As reported in Table 7.1.1 of Arby (2008), (real) GDP estimates of Pakistan Bureau of Statistics after 1999-2000 are over estimated by around 1 percent compared to Arby (2008). We have quarterised the PBS released GDP (Factor Cost) for 1999-2000 and onward.

Those who want to make use of previous (1971-72 to 1998-99) as well as our estimates (1999-00 to 2009-10) of quarterly GDP they can use the table (Appendix C) showing quarterly proportions of GDP for 1971-72 to 1998-99 based on Arby (2008) and for 2000-01 to 2009-10 based on this study. Combining the two studies‟ estimates is convenient as for the year 1999 -00 the quarterly proportions of real GDP are same from both the studies up to two decimal places.

This approach can be used in case of other sectors and subsectors of the GDP to make combined datasets for 1972-73 to 2009-10.

The overall estimated quarterly gross value added by services sector (at 1999-2000 prices) is shown below in figure 3. In figure 4, we have presented the time series plot of overall real quarterly GDP (FC).

11 Periodic budgetary increases in salary are taken care in annual figures and do not affect the monthly distribution of wages and salary

11

As observed, the seasonal patterns of services sector are closest to those of overall GDP. The same thing can be observed by analyzing the seasonal dispersion worked out in Table I2 in the Appendix.

4. Methodology-Expenditure Side

For quarterisation of expenditure side of the annual national income components for 1972-73 to 2009-10 we assume production side quarterly GDP equals the sum of quarterly components of national income. This assumption helps us finding quarterly private consumption expenditures as a residual

once we estimate quarterly government consumption expenditures,

12 In the (annual) National Income Accounting System of Pakistan private consumption is actually treated as residual (see Baqai, 1963).

400000 450000 500000 550000 600000 650000 700000 750000 800000 850000 900000

Q1-FY00 Q3-FY00 Q1-FY01 Q3-FY01 Q1-FY02 Q3-FY02 Q1-FY03 Q3-FY03 Q1-FY04 Q3-FY04 Q1-FY05 Q3-FY05 Q1-FY06 Q3-FY06 Q1-FY07 Q3-FY07 Q1-FY08 Q3-FY08 Q1-FY09 Q3-FY09 Q1-FY10 Q3-FY10

M i l l i o n R u p e e s

Figure 3: GVA of Services - Quarterly (1999-00 Prices)

700000 800000 900000 1000000 1100000 1200000 1300000 1400000 1500000 1600000

Q1-FY00 Q3-FY00 Q1-FY01 Q3-FY01 Q1-FY02 Q3-FY02 Q1-FY03 Q3-FY03 Q1-FY04 Q3-FY04 Q1-FY05 Q3-FY05 Q1-FY06 Q3-FY06 Q1-FY07 Q3-FY07 Q1-FY08 Q3-FY08 Q1-FY09 Q3-FY09 Q1-FY10 Q3-FY10

M i l l i o n R u p e e s

Figure 4: Gross Domestic Product (FC)- Quarterly (1999-00 Prices)

12

quarterly overall (private as well as government) investment and imports and exports of goods and services. We can quarterise the annual components of GDP using either direct or indirect way. We use direct way when we have basic data for quarterly (or even at higher) frequency as in the case of exports and imports of goods and services. We use indirect way, like the use of relevant indicators, when we do not have required data available at quarterly frequency as in the case of quarterly public sector investment expenditures prior to Q1FY02. In the following we explain the procedure used to quarterise these annual components of GDP.

4.1 Investment

Quarterly gross fixed capital formation (GFCF) has been estimated for 1972 to 2006 by Arby and Batool (2007). Using commodity flow approach Arby and Batool (2007) disaggregated GFCF into four components namely: machinery and equipment (M,), furniture and fixture (R), land improvement (L) and structure (S). Machinery and equipment is further divided into imported machinery and equipment (MF) and domestic machinery and equipment (MD).That is

𝐺𝐹𝐶𝐹

= 𝑀

+ 𝑅

+ 𝐿

+ 𝑆

; 𝑤ℎ𝑒𝑟𝑒 𝑀

= 𝑀𝐹

+ 𝑀𝐷

These components are then quarterised using combination of direct and indirect way. For imported machinery and equipment we already have quarterly imports under the heading „capital imports‟ from monthly bulletin of PBS. All other components are quarterised using indirect way that is on the basis of (quarterly) seasonality in the relevant indicator for which quarterly data is available. Following table shows the components, relevant indicators and their data sources. This last column of this table shows the deflator used for converting current prices investment components data into1999-2000 prices data.

Investment Component

Relevant Indicator High Frequency Data

Source

Deflator

MF Quarterly imported machinery and equipment

Monthly Bulletin of PBS Unit values index of imports

MD Quarterly LSM „machinery and equipment‟

sub-index (estimated by using relevant items and their weights)

Monthly Bulletin of PBS WPI manufacturing

S Quarterly LSM „building material‟ sub- index

Monthly Bulletin of PBS WPI building material

L Quarterly valued added of crops (with two quarters lead - assuming investment activity is undertaken before/during sowing of crops)

Arby (2008) and this study WPI food

R Five percent of quarterly „structure‟ (as is the case with annual NIA estimates)

This study WPI building material

Source: Arby and Batool (2007)

13

The quarterly GFCF is then estimated by summing the quarterly values of all these components.

We estimated the GFCF both based on current prices and based on 1999-2000 prices. It helped us to estimate the overall investment deflator.

4.1.1 Government Investment Expenditures

The annual estimates of GFCF in Pakistan are primarily constructed separately for private and government sectors. Problem with the Arby and Batool (2007) is that they quarterised overall GFCF rather than quarterisation of private and government investment separately.

However, sometimes, researchers are in need of analysing the private and public sector investment behaviour separately. In addition to estimating the overall investment, we have also attempted to quarterise the government sector investment separately as we explain here in this subsection.

Government investment is measured as total development expenditure (TDE) by the government. TDE includes the expenditures on public sector development program (PSDP) and other development expenditures. Quarterly TDE data is available from the fiscal operations statistics of ministry of finance, government of Pakistan, since the first Q1FY02. The average quarterly proportion of Q1FY02 to Q4FY10 of TDE has been used to quarterise the annual public sector investment for FY73 to FY01.

Year

Actual Quarterly Shares of TDE Jul-Sep Oct-Dec Jan-Mar Apr-Jun 2001-02 10.9223 30.2384 32.8802 25.9590 2002-03 16.7900 23.0352 23.9633 36.2115 2003-04 15.4791 19.8790 17.5881 47.0538 2004-05 14.0419 21.7888 24.7082 39.4611 2005-06 13.8980 21.2423 20.7927 44.0670 2006-07 15.5550 18.8592 21.8977 43.6880 2007-08 24.5615 24.3882 23.4411 27.6093 2008-09 12.6446 16.8657 24.6720 45.8178 2009-10 17.7678 18.8569 20.0187 43.3566 Average 15.74002 21.68374 23.32911 39.24712

Source: authors‟ calculations

We have used overall quarterly investment deflator (estimated as in section 4.1) to convert

quarterly nominal government investment expenditures into quarterly government investment

expenditures based on 1999-2000 prices.

14 4.1.2 Private Sector Investment

Subtracting quarterly public sector investment estimated as in section 4.1.1 from the overall quarterly investment estimated as discussed in section 4.1 we obtain the quarterly private sector investment.

The overall estimated quarterly gross fixed capital formation (at 1999-00 prices) is shown below

:

4.2 Imports and Exports of Goods and Services

Imports of goods and services are the components where we use the direct way of estimating the quarterly data from monthly information. Data on merchandise imports and exports are available from PBS in local currency (Pak Rupee). Data on non-factor services and other current transfers are obtained from SBP‟s monthly balance of payments statistics in US$

which are converted into local currency (Pak Rupee) by using local currency per US$ (average) exchange rate. Quarterly imports and exports of goods and services are reported in Appendix E.

4.3 Government Consumption Expenditures

To estimate the quarterly government consumption expenditures (GCE), the expenditures under General Public Service (GPSE) has been used as an indicator of GCE. The quarterly data is available for GPSE Q1FY02. The GPSE comprises of interest payments (domestic as well as

13 One may be interested in knowing why there was stagnancy (fall) in investment in late 1990s (late 2010s) and what were the implications for income and thus consumption. Answering such questions need a thorough investigation and is out of the scope of this study.

40000 90000 140000 190000 240000 290000

Q1-FY73 Q1-FY74 Q1-FY75 Q1-FY76 Q1-FY77 Q1-FY78 Q1-FY79 Q1-FY80 Q1-FY81 Q1-FY82 Q1-FY83 Q1-FY84 Q1-FY85 Q1-FY86 Q1-FY87 Q1-FY88 Q1-FY89 Q1-FY90 Q1-FY91 Q1-FY92 Q1-FY93 Q1-FY94 Q1-FY95 Q1-FY96 Q1-FY97 Q1-FY98 Q1-FY99 Q1-FY00 Q1-FY01 Q1-FY02 Q1-FY03 Q1-FY04 Q1-FY05 Q1-FY06 Q1-FY07 Q1-FY08 Q1-FY09 Q1-FY10

M i l l i o n R u p e e s

Figure 5: Quarterly Real Total Gross Fixed Capital Formation

15

external debt servicing), pensions, grant to non-government (total grants/others), and other general public services. For 2001-02 to 2009-10 the quarterly shares of GPSE have been used to quarterise the annual data of GCE. Before Q1FY02 there was no such availability of data on GPSE by the government of Pakistan. For pre Q1FY02 we need to find another indicator which correlates well with the quarterly pattern of GPSE. From the table (below) we can see that the quarterly pattern of GPSEs are very much close to the quarterly pattern of government overall tax revenues – lowest in the first quarter and highest in the last quarter while almost one-fourth in the middle two quarters.

Year

Share of GPSE Share of Tax Revenues

Jul-Sep Oct-Dec Jan-Mar Apr-Jun Jul-Sep Oct-Dec Jan-Mar Apr-Jun 2001-02 23.7361 21.8636 20.0619 34.3383 19.7085 23.9859 23.4805 32.8251 2002-03 23.1520 25.1034 24.7400 27.0046 20.1258 24.9209 22.9249 32.0284 2003-04 21.2113 26.5369 22.4539 29.7979 18.8727 25.5820 23.7300 31.8153 2004-05 20.2012 28.0886 25.5928 26.1175 21.2285 23.7306 23.5904 31.4506 2005-06 19.2950 29.3945 23.4479 27.8625 21.0408 24.5456 23.2561 31.1575 2006-07 16.9428 25.5767 27.1540 30.3265 21.5380 27.1813 22.5139 28.7668 2007-08 18.0272 26.9403 22.3037 32.7288 20.5177 22.3786 24.7881 32.3156 2008-09 20.3422 25.1254 24.3065 30.2259 22.9782 25.0011 22.5129 29.5077 2009-10 22.0936 23.4190 24.7729 29.7146 20.2864 24.3717 24.2302 31.1117 Average 20.5557 25.7832 23.8704 29.7907 20.6996 24.6331 23.4475 31.2198

Source: authors‟ calculations

Thus, we can use the quarterly pattern of government overall tax revenues to quarterise the annual GCE for pre Q1FY02 period. We have converted current price quarterly GCE into quarterly GCE based on 1999-2000 using consumer price index.

4.4 Private Consumption Expenditures

Quarterly private consumption expenditures (PCE) are obtained as residual by subtracting the sum of quarterly GCE, overall investment and net exports from the (production side) quarterly GDP provided in Appendix D.

The overall estimated quarterly real consumption is shown below in figure 6. In figure 7,

we have presented the time series plot of overall real quarterly GDP (FC).

16 5. Analysis of Results

It has been found that on average 21.73 percent of the real GDP was produced in the first quarter during 1999-2000 to 2009-2010 (See Table I1 in the Appendix I). It has been the lowest share of a quarter in annual real GDP. Second lowest share was observed in the third quarter around 24.43 percent. Highest quarterly share was observed in the second quarter which on average was found to be 27.16 percent. Second highest share was observed in the fourth quarter around 26.68 percent. These results are in line with the findings of Arby (2008) wherein quarterly shares for first, second, third and fourth quarters were found to be 21.8, 26.9, 25.2 and

175000 375000 575000 775000 975000 1175000 1375000

Q1-FY73 Q2-FY74 Q3-FY75 Q4-FY76 Q1-FY78 Q2-FY79 Q3-FY80 Q4-FY81 Q1-FY83 Q2-FY84 Q3-FY85 Q4-FY86 Q1-FY88 Q2-FY89 Q3-FY90 Q4-FY91 Q1-FY93 Q2-FY94 Q3-FY95 Q4-FY96 Q1-FY98 Q2-FY99 Q3-FY00 Q4-FY01 Q1-FY03 Q2-FY04 Q3-FY05 Q4-FY06 Q1-FY08 Q2-FY09 Q3-FY10

M i l l i o n R u p e e s

Figure 6: Quarterly Overall Real Consumption

200000 400000 600000 800000 1000000 1200000 1400000 1600000

Q1-FY73 Q2-FY74 Q3-FY75 Q4-FY76 Q1-FY78 Q2-FY79 Q3-FY80 Q4-FY81 Q1-FY83 Q2-FY84 Q3-FY85 Q4-FY86 Q1-FY88 Q2-FY89 Q3-FY90 Q4-FY91 Q1-FY93 Q2-FY94 Q3-FY95 Q4-FY96 Q1-FY98 Q2-FY99 Q3-FY00 Q4-FY01 Q1-FY03 Q2-FY04 Q3-FY05 Q4-FY06 Q1-FY08 Q2-FY09 Q3-FY10

M i l l i o n R u p e e s

Figure 7: Quarterly Real GDP at Factor Cost

17

26.1 percent respectively. As against the dispersion between the quarterly shares in real GDP, half yearly shares were found to be closer (in the ratio of 48.89:51.11) during FY00 to FY11. We have attempted to see in detail the sources of dispersion in the quarterly shares in real GDP. For this purpose, we have calculated dispersion in quarterly shares of various sectors and subsector of GDP (see Table I2 in the Appendix I). Highest dispersion is found to be in the quarterly shares of major crops. Maximum (at 41.37 percent) and minimum (10.29 percent) of quarterly shares are found to be for major crops. This observation is in line with the findings of Arby (2008).

Second highest dispersion is found to be in the quarterly shares of finance & insurance subsector.

Probably, it is the major crops which are driving the seasonality in the agricultures sector and in finance & insurance, and even to the overall real GDP. This can be said on the basis of fact that (net) operating surplus of financial institutions and insurance companies has been quarterised in this study using the seasonal factors of broad money supply (M2). Seasonality in M2 itself is mainly driven by seasonality in agriculture business credit (see Figure 8, which shows seasonal factors of M2, private agriculture business credit (ABC) and private non-agriculture business credit (NABC) during January 2004 to June 2010 for which we have disaggregated data available).

We also investigated, if (during FY2000 to FY2010) the seasonal component in quarterly gross value added (QGVA) by agriculture drives the seasonal components extracted from the QGVA of industrial sector, QGVA of services sector and the overall GDP (FC). Seasonality in industry, services and overall GDP is found to be caused by (in Granger (1969) sense) the seasonality in agriculture sector.

If we analyze the quarterly estimates of various components of expenditures side of the NIAs for FY1973-FY2010 we observe that it is the real investment which is more volatile

in

14 Measured by four quarter coefficient of variation. It does not make any difference to our findings even if we use 4 quarters rolling coefficient of variation. The same is observed if we use the 4 quarters rolling coefficient of variation upon the cyclical components extracted from seasonally adjusted real consumption and income.

-0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4

-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Jul-Sep Oct-Dec Jan-Mar Apr-Jun

Figure 8: Seasonal Patterns in Money Supply, Private Agriculture Based Credit and Private Non ABC

M2 NABC ABC (RHS)

Source: authors' calculations

18

Pakistan, followed by consumption and income (see Figure 9). The same is also true if we compare the volatility of the cyclical components of real consumption, investment, and income (see Table I3 in the Appendix). We also observed this phenomenon when we compared the variances of the growth rates in the per capita

real consumption, investment and income.

Choudhary and Pasha (2013) also found similar results but for annual frequency. Our findings are in line with the observation by Aguiar and Gopinath (2007) at levels and with Gracia et al (2010) at growth.

In order to see whether the moments of annual and quarterised data match, we compared the important moments (like variances and correlation coefficients) calculated from the cyclical components of real consumption, investment, and income as reported in Table I3 in the Appendix. The closeness

of these estimated moments gives us due comfort that our quarterised data can be used in business cycle research.

15 Annual data on population is Pakistan is used from World Bank Development Indicators website (retrieved on February 21, 2013) . It is converted in to quarterly population using EViews 7 based on „linear match last‟ approach.

16 The correlation coefficient between the cyclical components of real investment and the real income from observed annual and quarterised data may be different because in the case of Pakistan the government investment crowds in the private investment, and the government investment is intrinsically volatile in Pakistan and thus generates larger (than usually observed) volatility in overall investment in the country during the year, which disappears on aggregation.

0 5 10 15 20 25

1972-73 1973-74 1974-75 1975-76 1976-77 1977-78 1978-79 1979-80 1980-81 1981-82 1982-83 1983-84 1984-85 1985-86 1986-87 1987-88 1988-89 1989-90 1990-91 1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10

P e r c e n t

Figure 9: Volatility (Coefficient of Variation) in consumption, investment, and income

CV (Quarterly Overall Real Consumption)

CV (Quarterly Total Gross Fixed Capital Formation) CV (Quarterly Real GDP at Factor Cost)

19 6. Concluding Remarks

Pakistan Bureau of Statistics (PBS) compiles NIAs for Pakistan on annual basis. Despite availability of some data on higher frequency like for large scale manufacturing output and merchandise trade, PBS does not provide NIA on quarterly basis which contains a comprehensive picture of economy than provided by the indicators like LSM index. In order to meet needs of the researchers some studies in the past provided estimates of quarterly GDP but from production side only. These studies include Bengaliwala (1995), Kemal and Arby (2004), and Arby (2008). None of these studies provide estimates of quarterly GDP from expenditure side.

Arby (2008) provides the quarterly estimates of the annual GDP from production side (for 1972 to 2005) based on current as well as on current (1999-2000) prices. Realizing the need to provide quarterly production estimates for the last decade, this study provides quarterly estimates of (sectoral and overall) gross domestic production in Pakistan during 1999-2000 to 2009-2010 based on constant prices of 1999-2000 as well as on current prices following Arby (2008). Quarterly GDP production side estimates for 1973-1999 based upon Arby (2008) and those for 2000-2010 based upon this study are combined in here to have a longer QGDP from production side. Then we quarterise the expenditures side of the NIAs. This is first study which provides estimates of various components of expenditures side of Pakistan‟s GDP for 1972 -1973 to 2009-2010. On the production side, we find that the agriculture sector is the main driver of (quarterly) seasonality in Pakistan‟s NIA statistics. On the expenditure side, we observed that real investment is the most volatile in Pakistan followed by consumption and income. Matching of moments of some core macro variables for the annual and quarterised data suggests that the estimated quarterly data provided in this study can be used in research.

This will be even useful for assessing, analyzing and monitoring the ongoing state and

performance of economy by the policy makers if PBS (herself) starts providing the estimates of

QNIA on regular basis. Particularly, in the environment of high and variable inflation, QNIA are

quite important from policy point of view as in these circumstances one of the assumptions of

ANIA (price homogeneity during the year) is violated and summing the current prices data over

a year may be less useful.

20

Appendix A: Quarterly GDP – Production Side (Real) for 1999-00 to 2009-10

Gross Domestic Product- Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 768920 983290 873797 936010 3562018

2000-01 807466 977510 897929 949186 3632091

2001-02 829474 1011540 917400 986704 3745118

2002-03 857833 1050535 974514 1039222 3922104

2003-04 919079 1132786 1045871 1117846 4215582

2004-05 983493 1256764 1124251 1228723 4593230

2005-06 1041520 1331002 1177106 1310848 4860476

2006-07 1113882 1410238 1255378 1412211 5191709

2007-08 1164522 1467331 1310092 1441067 5383012

2008-09 1187247 1492680 1304320 1491470 5475716

2009-10 1227898 1534075 1369976 1511653 5643602

Gross Value Added of Agriculture – Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 176572 282214 198884 265938 923609

2000-01 175820 278982 197187 251510 903499

2001-02 175766 279418 202327 246922 904433

2002-03 180757 288148 210502 262535 941942

2003-04 186331 296877 215772 265847 964827

2004-05 196641 325611 217846 287305 1027403

2005-06 212282 347883 236960 294973 1092098

2006-07 214773 353830 251170 317264 1137037

2007-08 221505 361354 263807 302185 1148851

2008-09 227244 371884 262679 333196 1195002

2009-10 230577 380861 266650 324323 1202411

Gross Value Added of Major Crops - Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 34469 117955 42346 147430 342200

2000-01 31870 110213 36649 129742 308474

2001-02 30854 107350 39324 123383 300911

2002-03 32190 111180 42783 135352 321505

2003-04 32739 115139 43501 135678 327057

2004-05 41397 142029 44262 157370 385058

2005-06 40544 140017 40865 148579 370005

2006-07 40430 140728 50273 167187 398617

2007-08 37339 137963 52837 145049 373188

2008-09 40364 142875 46777 172119 402135

2009-10 41638 145740 45625 159980 392983

21

Gross Value Added of Minor Crops- Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 49778 24649 28199 23054 125679

2000-01 47991 23798 27262 22622 121673

2001-02 46174 22972 26175 21896 117217

2002-03 46858 23478 26741 22369 119446

2003-04 49302 24285 27746 22788 124121

2004-05 50370 24834 28369 22419 125993

2005-06 50799 24606 28581 22471 126457

2006-07 49973 24423 28387 22460 125243

2007-08 55065 27325 31808 24689 138887

2008-09 53937 27076 31436 24752 137201

2009-10 50451 24796 28943 22418 126609

Gross Value Added of Live Stocks - Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 85510 129307 116794 85510 417120

2000-01 88779 134250 121258 88779 433066

2001-02 92038 139180 125711 92038 448968

2002-03 94401 142753 128939 94401 460495

2003-04 97118 146861 132649 97118 473745

2004-05 99400 150312 135765 99400 484876

2005-06 115108 174065 157220 115108 561500

2006-07 118367 178994 161672 118367 577400

2007-08 123289 186436 168394 123289 601408

2008-09 127152 192278 173671 127152 620253

2009-10 132591 200503 181099 132591 646783

Gross Value Added of Fishing - Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 2105 4299 4789 3970 15163

2000-01 2043 4172 4647 3853 14715

2001-02 1791 3658 4074 3378 12901

2002-03 1853 3784 4215 3495 13346

2003-04 1890 3859 4298 3564 13611

2004-05 1901 3882 4324 3585 13691

2005-06 2296 4689 5223 4331 16540

2006-07 2649 5410 6026 4996 19080

2007-08 2892 5907 6580 5455 20834

2008-09 2960 6044 6733 5582 21319

2009-10 3003 6133 6832 5664 21632

22

Gross Value Added of Forestry - Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 4711 6005 6757 5974 23447

2000-01 5137 6549 7370 6515 25571

2001-02 4909 6258 7042 6226 24436

2002-03 5454 6953 7825 6918 27150

2003-04 5282 6734 7578 6699 26293

2004-05 3573 4555 5126 4532 17785

2005-06 3535 4506 5071 4483 17596

2006-07 3354 4276 4812 4254 16697

2007-08 2920 3722 4189 3703 14534

2008-09 2831 3609 4062 3591 14094

2009-10 2894 3689 4151 3670 14404

Gross Value Added of Industry - Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 184516 223680 222827 199840 830863

2000-01 201834 212577 229531 221254 865196

2001-02 214400 214535 230355 229249 888539

2002-03 211024 220866 250551 243742 926183

2003-04 242604 260501 289265 284439 1076808

2004-05 269031 297285 329424 311528 1207268

2005-06 287049 302711 335352 331715 1256827

2006-07 321990 326136 360230 359176 1367532

2007-08 329948 334712 365099 357358 1387117

2008-09 341416 332955 354347 356950 1385669

2009-10 346718 356219 387304 379869 1470110

Gross Value Added of Mining – Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 19269 20742 20624 20415 81050

2000-01 20387 22526 20983 21632 85528

2001-02 22596 22354 22039 23442 90431

2002-03 22980 23618 24405 25414 96418

2003-04 25186 27491 29462 29334 111473

2004-05 27769 28774 36531 29547 122621

2005-06 29771 32441 32880 33197 128288

2006-07 31931 32972 33647 33703 132254

2007-08 31239 41512 30680 34616 138047

2008-09 36928 30400 33199 36820 137348

2009-10 36123 33170 33764 37357 140415

23

Gross Value Added of Large scale manufacturing - Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 71196 93278 95970 78158 338602

2000-01 85298 92230 103563 94596 375687

2001-02 90511 92789 105441 100117 388859

2002-03 92118 98223 117961 108653 416955

2003-04 105999 118615 138233 129785 492632

2004-05 126004 146204 162364 156187 590759

2005-06 143252 152044 172298 171992 639585

2006-07 161160 163081 184889 186358 695489

2007-08 172459 166832 195391 188944 723626

2008-09 162631 160994 171578 170082 665285

2009-10 161899 167997 186973 180279 697148

Gross Value Added of Small scale manufacturing- Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 38730 50743 52208 42518 184199

2000-01 44426 48036 53939 49268 195670

2001-02 48410 49629 56395 53548 207982

2002-03 48845 52082 62548 57613 221089

2003-04 50523 56536 65887 61860 234807

2004-05 40551 47052 52253 50265 190121

2005-06 46286 49127 55671 55572 206656

2006-07 51759 52376 59380 59851 223365

2007-08 57231 55364 64842 62702 240139

2008-09 63111 62476 66583 66003 258173

2009-10 64458 66886 74441 71776 277562

Gross Value Added of Slaughtering - Quarterly

(1999-2000 prices) (Rs Millions)

Years\Quarters Jul-Sep Oct-Dec Jan-Mar Apr-Jun Annual

1999-00 n.a n.a n.a n.a n.a

2000-01 n.a n.a n.a n.a n.a

2001-02 n.a n.a n.a n.a n.a

2002-03 n.a n.a n.a n.a n.a

2003-04 n.a n.a n.a n.a n.a

2004-05 10685 14841 20777 13060 59363

2005-06 12008 16678 23349 14677 66712

2006-07 12500 17362 24306 15278 69447

2007-08 13020 18084 25318 15914 72336

2008-09 13570 18847 26386 16585 75388

2009-10 14157 19663 27528 17303 78652